Data Sensemaking Framework by Applying Conceptual Modeling and Data Analysis: A Case Study from the Norwegian Public Transportation Domain

ABSTRACT The industry aims to adopt data‐driven methods to improve system availability and reliability. A critical challenge in this process is achieving data sensemaking for practitioners, including maintenance personnel and decision‐makers, product developers, and consultants. This article replicates a data framework that combines conceptual modeling and data analysis to achieve data sensemaking. The main author previously developed and exemplified the framework through a case study of an automated parking system in a small‐and‐medium‐sized enterprise. This study implemented the framework in a more complex context: the Norwegian maintenance Metro public transportation department within a large enterprise, using Oslo Metro. Applying the framework assisted the maintenance department in closing the internal maintenance feedback loop. In this context, we collected and analyzed exploratively internal failure data and external weather data spanning 6 years. In parallel, we developed several conceptual models co‐created with various stakeholders. These models facilitated a shared understanding of the maintenance processes and guided exploratory data analysis, with data analysis, in turn, supporting conceptual modeling. This study implemented and evaluated the framework using on‐site and participant observations, interviews, subject matter experts’ feedback, and co‐creation sessions. Findings from the observations emphasize the complementary nature of conceptual modeling and data analysis in achieving data sensemaking, leading to a deeper understanding of Oslo Metro behavior. This understanding aids in improving maintenance processes and their effectiveness, enhancing Oslo Metro availability and reliability. Despite the framework implementation is context‐dependent, findings indicate its effectiveness, motivating practitioners to implement it in other contexts.